PROJECT SUMMARY Double-stranded (ds) RNA is an early danger signal that alerts the host to viral invasion and activates several innate immune pathways that limit virus replication. These pathways include type I and type III interferons that induce antiviral effectors, the oligoadenylate synthetase-ribonuclease L (OAS-RNase L) system that degrades ssRNA and leads to apoptotic death, and the protein kinase RNA dependent (PKR) pathway that halts protein synthesis. These host responses to dsRNA, and the many mechanisms viruses use to subvert the effector pathways, have been studied extensively in RNA viruses. However, there is a relative dearth of studies on dsRNA production during infection by DNA viruses, and there is a gap in our understanding of downstream effects or viral antagonism of antiviral pathways. Adenovirus (AdV) has a double-stranded DNA genome that has served as a powerful system for seminal discoveries in RNA biology, aided by availability of genetic mutants. AdV and other viruses with limited genome size and protein coding capacity have evolved to maximize gene expression through regulated transcription and use of both DNA strands for protein production. Thus, annealing of complementary single-stranded RNAs produced by symmetrical transcription of DNA virus genomes could lead to dsRNA. Although AdV is known to counter IFN responses and block PKR activation, it is not known whether infection generates dsRNA and how the antiviral dsRNA-activated pathways are evaded in infected cells. AdV mutants with early regions deleted have been useful for deciphering key viral functions required for infection. Deletion of early E1B and E4 genes results in unstable viral RNAs that are poorly transported and translated. The E1B55K and E4orf6 gene products form an E3 ubiquitin ligase required for efficient virus production. However, there is another gap in our understanding of how ubiquitination of substrates by this complex promotes RNA processing and late viral protein synthesis. We recently discovered that infection with AdV mutants that are defective for E1B55K or E4 generates dsRNA that accumulates in the nucleus, and that RNase L and PKR responses are activated. We also showed that a functional E1B55K/E4orf6 complex is required to prevent dsRNA during AdV infection and we have identified cellular RNA binding proteins that are ubiquitinated by the viral complex. These preliminary results have led to our overall hypothesis that the activity of the viral ligase ubiquitinates cellular RNA processing factors to prevent accumulation of dsRNA during infection and overcome antiviral host responses. Our Specific Aims are 1) to identify the source of dsRNA (viral or host), determine host responses activated, and 2) define how the E1B55K/E4orf6 ubiquitin ligase activity prevents antiviral responses to dsRNA. In this way we will use AdV as a model pathogen to study how dsRNA responses impact infection for DNA viruses. Our long-term goal is to uncover fundamental principles of gene expression by deciphering how DNA viruses manipulate RNA biogenesis pathways and evade antiviral defenses.